The O'Reilly Finance ML Book, But Runnable

What it does This repository contains the Jupyter notebooks for the O’Reilly book Machine Learning and Data Science Blueprints for Finance. It provides runnable versions of case studies that apply supervised learning, reinforcement learning, NLP, and clustering to financial domains including algorithmic trading, portfolio management, and robo-advisors.

The interesting bit Rather than organizing by chapter number, the README groups notebooks by financial application—trading strategies, portfolio management, and robo-advisors—so you can jump straight to the domain you care about without mapping book sections to repository folders.

Key highlights

• Case studies include Bitcoin trading via classification, dimensionality reduction for trading speed, clustering for pairs trading, reinforcement learning strategies, and NLP sentiment analysis
• Covers portfolio management with eigen portfolios and robo-advisor construction via supervised regression
• Offers cloud execution through Binder, Google Colab, and static browsing via nbviewer
• Organized by financial application rather than by textbook chapter

Caveats

• The README is dominated by environment setup instructions and reveals little about the actual methodology or performance of the case studies
• Several notebook descriptions are truncated mid-sentence in the README, and the authors explicitly warn that data file paths break in Google Colab and require manual URL substitution
• Context and explanations live in the book; the repo is strictly the executable companion

Verdict Useful if you own the O’Reilly text and want to run or tweak the examples. If you’re seeking a self-contained finance ML framework or detailed documentation independent of a paid book, this won’t satisfy.